A Ligand-Enabled Palladium-Catalyzed Highly para-Selective Difluoromethylation of Aromatic Ketones

Article abstract of DOI:10.1002/anie.201809788

A practical and highly para-selective C?H difluoromethylation of aromatic ketones has been developed by employing tetrakis(triphenylphosphine)palladium(0) as the catalyst and triphenylphosphine as the ligand. In addition to general aromatic ketones, this

Full text of DOI:10.1002/anie.201809788

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: A Ligand Enabled Palladium Catalyzed Highly para-Selective
Difluoromethylation of Aromatic Ketones
Authors: Guangliang Tu, Chunchen Yuan, Yuting Li, Jingyu Zhang,
and Yingsheng Zhao
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.201809788
Angew. Chem. 10.1002/ange.201809788
Link to VoR: http://dx.doi.org/10.1002/anie.201809788
http://dx.doi.org/10.1002/ange.201809788

10.1002/anie.201809788
Angewandte Chemie International Edition
COMMUNICATION
A Ligand Enabled Palladium Catalyzed Highly para-Selective
Difluoromethylation of Aromatic Ketones
Guangliang Tu,† Chunchen Yuan,† Yuting Li,† Jingyu Zhang,† and Yingsheng Zhao*†
Dedicated to Professor Xiyan Lu on the occasion of his 90th birthday
Abstract:
A
practical
and
highly
para-selective
C–H
challenging11-13 and requires an urgent solution. Herein, we
difluoromethylation of aromatic ketones has been developed by
employing tetrakis(triphenylphosphine)palladium(0) as the catalyst
and triphenylphosphine as the ligand. In addition to general aromatic
ketones, this transformation was compatible with bioactive
compounds and well-known drugs, such as oxybenzone, ketoprofen,
zaltoprofen, and propafenone. Moreover, the mechanistic study
revealed that the palladium complex coordinated with the carbonyl
group to affect the highly para-selective difluoromethylation reaction.
report
our
new
discovery
(Figure
1b)
that
tetrakis(triphenylphosphine)palladium(0) catalyzed the highly
para-selective direct difluoromethylation of aromatic ketones
with great substrate scope and functional group tolerance.
Aromatic ketones are important scaffolds that occur in a
wide range of biologically active molecules and drugs, such as
oxybenzone, ketoprofen, zaltoprofen, and propafenone. On the
other hand, the presence of the difluoromethylene group in a
compound can enhance the metabolic stability, conformational
changes and dipole moments, along with increased acidity of
their neighboring groups1. Therefore, the development of new
methodologies for achieving direct site-selective C–H
functionalization of aromatic ketones, especially through the
direct introduction of the difluoromethylene group, is highly
attractive for the discovery of new drugs2.
Figure 1. The direct C–H difluoromethylation arenes.
Recently, Nakao et al. reported a pioneering work on the
para-selective alkylation of benzamides, diaryl ketones and aryl
sulfones by using nickel as the catalyst and a bulk aluminum
complex as the co-catalyst14. Inspired by the aluminum Lewis
acid-enabled highly para-selective functionalization of arenes,
we envisaged that the site-selective difluoromethylation of
aromatic ketones could be achieved by the coordination of a
suitable Lewis acid with the carbonyl group of the ketone.
In the last few decades, various approaches to directly
introduce the functionalized fluorinated moieties on aromatic
rings have been extensively explored. For example, in 2001, the
group of Chen has developed a groundbreaking fluoroalkylation
reaction of the electronic-rich aromatic compounds with
perfluoroalkyl chlorides in the presence of sodium dithionite3.
Later, the research groups of Fuchigami4, Yamakawa5, Shen6,
and others7 conducted extensive studies on the direct
introduction of fluorine-containing functional groups, especially
those containing difluoromethylene (CF2) groups, on aromatic
rings. Although heteroaromatic compounds are capable of
achieving site-selective fluorinated products due to their intrinsic
electronic effects3,7a,d, the general aromatic compounds usually
suffer from poor site-selectivity7c,e,f (Figure 1a). Till recently, the
site selective C-H difluoromethylation of arenes have achieved a
preliminary progress8-10. For example, the group of Ackermann
Initially,
acetophenone
(1a)
was
treated
with
bromodifluoroacetate (2)8,15, a cheap and commercially available
difluoromethylation reagent, in the presence of transition
metals16, and various Lewis acids. Unfortunately, a mixture of
the para- and meta-difluoromethylated products (3a and 4a,
respectively; para/meta (p/m) ≈ 6/1 to 10/1) was obtained in less
than 20% yield (see SI). Even though palladium acetate gave
the difluoromethylated product in 27% yield, the site selectivity
was still poor (Table 1, entry 1). The ligands 2,2'-bipyridine, Ac-
Gly-OH and DPEphos17 all provided 3a in less than 5% yield,
along with the starting material 1a being recovered (entries 2–4).
Gratifyingly,
triphenylphosphine
afforded
the
para-
difluromethylated acetophenone 3a in 33% yield (Table 1, entry
5), whereas only a trace amount of the meta-difluoromethylated
product 4a was observed (p/m > 20/1). It is worth noting that the
strong coordinating ability of bidentate phosphorous ligands
(DPEphos) might prevent the palladium from coordinating with
the carbonyl group of ketone. When Pd(PPh3)4 was used as the
catalyst in the absence of a ligand, the product 3a was obtained
in 35% yield with high para-selectivity (entry 6). A slightly
improved yield was obtained with PPh3 as the ligand, leading to
the product 3a in 44% yield (entry 7). The large number of
bromine anions formed in the reaction might have coordinated
with the palladium catalyst and subsequently resulted in its
deactivation. However, in the presence of an excess amount of
PPh3 in the reaction, the better coordination ability of PPh3 than
that of bromine anions would prevent the latter from coordinating
with palladium catalyst18. Further screening revealed that
potassium acetate was the best base and n-hexane was the
firstly developed
a
ruthenium-catalyzed meta-selective C-H
para-
difluoromethylation of 2-arylpyridine derivatives8.
A
selective difluoromethylation of acetanilides and ketoximes by
modulating the coordination ability of the directing group was
developed by our group9. However, the direct para-selective
difluoromethylation of aromatic ketones still remains
[†]
G. Tu, C. Yuan, Y. Li, J. Zhang, Prof. Dr. Y. Zhao
Key Laboratory of Organic Synthesis of Jiangsu Province, College
of Chemistry, Chemical Engineering and Materials Science
Soochow University, Suzhou 215123 (PR China)
E-mail: yszhao@suda.edu.cn
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/anie.201809788
Angewandte Chemie International Edition
COMMUNICATION
best solvent, affording 3a in 76% yield and slightly improved
para-selectivity (entry 8). The addition of AgF gave a slight
promoting effect, affording the product 3a in 81% yield with
slightly reduced site-selectivity (p/m= 21/1; Table 1, entry 9).
Although the role of AgF is unclear, it is believed to act as a
Lewis acid-assisted palladium complex to activate the aromatic
ring. In addition, a control reaction revealed that the palladium
catalyst is indispensable for this difluoromethylation reaction.
was well tolerated, leading to the corresponding products in
good yields (6g, 63%; 6h, 61%). When 1,4-dioxane was
employed as the solvent, the substrates pyruvophenone (5i) and
Table 2. Functional group tolerance on acetophenone.[a]
Table 1. Optimization of reaction conditions.[a]
[a] Reaction performed in 0.2 mmol scale with 2 (1 mmol) in a sealed tube;
isolated yields.
[a] Reaction performed in 0.2 mmol scale with 2 (1 mmol) in a sealed tube. [b]
GC yield with tridecane as the internal standard. [c] The ratio of selectivity
determined by 1H NMR. [d] KOAc (4 equiv). [e] AgF (30 mol %). [f] Isolated
yield.
benzyl (5j) afforded the corresponding products in moderate
yields.
Methyl
3-benzoylpropionate
(5k),
methyl
4-
With optimized reaction conditions in hand, we next
explored the functional group tolerance on acetophenone and
the results were presented in Table 2. The ortho-substituted
acetophenones all performed well under the optimized reaction
conditions, providing the para-difluoromethylated products in
benzoylbutanoate (5l), butyrophenone (5m), valerophenone
(5n), and isobutyrophenone (5o) were all compatible, providing
the corresponding products in moderate to good yields (6k–6o).
The diphenyl ketone derivatives were also subjected to the
optimized reaction conditions to explore the functional group
tolerance. All the diphenyl ketones were compatible with this
reaction, leading to the corresponding products in moderate to
good yields (6p–6w). Moreover, various functional groups such
as Me, OMe, OBn, OAc, Br, CO2Me, and CF3 were well tolerated
on the diphenyl ketone scaffold. It was worth noting that the 3-
good
yields
(3b–3e).
Unfortunately,
only afforded
2-
the
(trifluoromethyl)acetophenone
1f
difluoromethylated product 3f in trace amounts, along with the
starting material being recovered. It is probable that the strong
electronic-withdrawing effect of the trifluoromethyl group reduces
the activity of the aromatic ring, resulting in only trace amounts
of difluoromethylated product. Impressively, the functional
groups such as Me, OMe, OBn, OCF3, Cl, and Br on the meta-
position of the aromatic ring were all well tolerated, affording the
highly para-selective difluoromethylated aromatic ketones (3g–3l)
in moderate to good yields (48%–81%).Further study revealed
that the disubstituted aromatic ketones (1m–1p) reacted well,
leading to the highly para-selective difluoromethylated products
in moderate to good yields (3m–3p; 43%–80%).
methoxybenzophenone afforded
a
mixture para-selective
difluoromethylated product (See SI). The substrate 5x reacted
well, selectively affording the product 6x in 74% yield. Both
diphenyl ketone (5p) and dibenzosuberone (5z) all provided the
mono-para-difluoromethylated products in good yields,
accompanied with bis-difluoromethylated products in less than
20% yields.
Fluorine-containing functional groups usually improve the
bioactivity of the drugs. Hence, directly introducing fluorine-
containing functional groups into natural bioactive compounds
could accelerate the development of new drugs. Delightfully, our
method could be successfully applied to the late-stage
functionalization of several well-known drugs such as ketoprofen,
zaltoprofen, oxybenzone and chrysazin to yield the
corresponding products (8a-d) in moderate to good yields, along
We next explored the diversity of the aromatic ketone
substrates under the optimized conditions and the results were
presented in Table 3. The benzocyclic ketones including 1-
tetralone (5a), 6-methoxy-1-indanone (5b), and chromanone
(5c) reacted smoothly with bromodifluoroacetate (2) to afford the
corresponding para-difluoromethylated products 6a–6c in
moderate to good yields (73%, 52% and 46%, respectively),
along with the unreacted starting material being recovered. All
the reactions with the substrates 5d–5f proceeded smoothly,
resulting in the para-difluoromethylated products in good yields.
Impressively, the presence of a hydroxy group on the substrates
with the starting material being recovered (Scheme 1).
A
difluoromethylated propafenone key precursor19 8e was also
obtained in moderate yield. We next explored the further
transformation of these difluoromethylated bioactive compounds.
For example, the benzyl group of the difluoromethylated product
This article is protected by copyright. All rights reserved.

10.1002/anie.201809788
Angewandte Chemie International Edition
COMMUNICATION
(8c) could be removed under a known reaction procedure in
excellent yield (9a)20. The difluoromethyl group could be readily
transformed into various functional groups (9b–9d).
the difluoromethylation was completely inhibited, indicating that
the reaction proceeds through a free radical pathway (see SI).
Second, when acetophenone was treated with the difluoromethyl
radical without any transition metal21, a mixture of para-and
meta-difluoromethylated acetophenones was obtained in overall
20% yield (p/m = 5/1; Scheme 2a). This experimental result was
consistent with the work reported by the Zhou research group17.
Third, the substrates of 10 provided single difluoromethylated
products in good yields with high selectivity at the para-position
of the benzoyl ring rather than the electronic rich aromatic ring
(Scheme 2b). These results indicated that the coordination of
free arene-bound ketone carbonyl group with the palladium
catalyst is the key factor in realizing the para-selective C–H
difluoromethylation. A primary kinetic isotope effect (1.0) was
observed under standard reaction conditions indicating that the
C–H bond activation might not be the rate-determining step.
Furthermore, the acetophenone-D5 (1a-[D5]) was used instead
of acetophenone under the standard conditions, which afforded
the difluoromethylated product 3a-[D] in 75% yield without any
D/H scrambling in NMR analysis (Scheme 2c). Meanwhile, when
7-methoxy-1-indanone (12), in which both of the ortho-positions
Table 3. Substrate scope of aromatic ketones.[a]
were blocked, was employed as
a substrate, the para-
difluoromethylated product 13 was obtained in 43% yield
(Scheme 2d). These outcomes indicated that the palladium
complex only coordinated with the carbonyl group of the
aromatic ketone, activating the carbonyl group-linked aromatic
ring without the formation of a five-membered palladacycle,
while the steric hindrance effect of the ligand triphenylphosphine
hinders the ortho and meta-difluoromethylation on the aromatic
ring, resulting in the high para-selectivity. A control experiment
was performed to support the hypothesis, such as when using 4-
methoxyacetophenone (1q) in which the para position is blocked
by a methoxy group, as a substrate (Scheme 2e), only trace
amount of the difluoromethylated product 3q and 95% of 4-
methoxyacetophenone were obtained.
[a] Reaction performed in 0.2 mmol scale with 2 (1 mmol) in a sealed tube;
Isolated yields. [b] 1,4-dioxane. [c] 120 oC, 1,4-dioxane.
Scheme 1. Late-stage functionalization of drugs and further transformations
To explore the mechanism of this palladium-catalyzed C–H
difluoromethylation reaction and to explain the basis for the
para-selectivity, various parallel reactions were performed
(Scheme 2). First, when equal amounts of the radical scavenger
TEMPO and bromodifluoroacetate were added in the reaction,
Scheme 2. Preliminary mechanistic study
This article is protected by copyright. All rights reserved.

10.1002/anie.201809788
Angewandte Chemie International Edition
COMMUNICATION
When the commercially available di-µ-bromobis(tri-t-
butylphosphino)dipalladium(I) was used as the catalyst instead
of Pd(PPh3)4, the difluoromethylated acetophenone 3a was
obtained in 52% yield with high para-selectivity (Scheme 3a).
When equal amounts of bromodifluoroacetate, acetophenone,
and [Pd(I)P(t-Bu)3Br]2 were employed in the reaction, the
product 3a was obtained in 25% yield, along with the recovery of
acetophenone (Scheme 3b). Interestingly, the product 3a was
not observed when Pd(PPh3)4 was used instead of [Pd(I)P(t-
Bu)3Br]2. However, an analogous result was obtained when two
equivalents of bromodifluoroacetate were used. Altogether,
these results indicated that Palladium(I) might be the key
catalytic species, which would oxidize to bromodifluoroacetate to
Various functional groups were well tolerated in the reaction,
affording the corresponding para-difluoromethylated products in
moderate to good yields. Moreover, the important
difluoromethylene group (CF2) was successfully introduced into
several well-known drugs such as oxybenzone, ketoprofen,
zaltoprofen, and propafenone. The mechanistic study indicated
that the palladium complex coordinated to the carbonyl group,
resulting in the highly para-selective difluoromethylation reaction.
Acknowledgements
This work was supported by Natural Science Foundation of
China (No.21772139, 21572149), Jiangsu Province Natural
Science Found for Distinguished Young Scholars (BK20180041)
Project of Scientific and Technologic Infrastructure of Suzhou
(SZS201708) and the PAPD Project.
generate
a
palladium(II) species and free radical of
difluoroacetate15, 22
.
Keywords: Site selectivity • Palladium • Ligand •
Difluoromethylation • Bioactive compounds
[1]
F. Robert, K. Yoshiro, Yagupolskii, Organofluorine Compounds in
Medicinal Chemistry and Biomedical Applications; Elsevier: New York,
1993.
[2]
a) D. O’Hagan, Chem. Soc. Rev. 2008, 37, 308; b) C. Ni, M. Hu, J. Hu,
Chem. Rev. 2015, 115, 765.
[3]
[4]
[5]
[6]
[7]
X.-T. Huang, Z.-Y. Long, Q.-Y. Chen, J. Fluorine Chem. 2001, 111, 107
S. Murakami, H. Ishii, T. Fuchigami, Tetrahedron 2006, 62, 3761.
Y. Ohtsuka, T. Yamakawa, Tetrahedron 2011, 67, 2323.
Scheme 3. Control experiments
Based on the above results and previous reports15, 22, a
plausible catalytic cycle was proposed in Scheme 7. A Pd(II)
intermediate was generated by the single electron transfer (SET)
between Pd(I) and bromodifluoroacetate, which further acted as
a weak Lewis acid coordinated to the carbonyl group in the
aromatic ketone to improve the reactivity of the carbonyl group
bound-arene. The generated difluoromethyl radical may be
directly coupled with complex I, producing the complex II,
followed by a SET process by the Pd(II) and aromatization to
generate the product 3a and Pd(I) catalyst.
S. Chen, C. Xu, L. Lu, Q. Shen, Chin. J. Chem. 2013, 31, 901.
a) M.-C. Belhomme, T. Poisson, X. Pannecoucke, J. Org. Chem. 2014,
79, 7205; b) H. Jiang, C. Huang, J. Guo, S.Yu, Chem. Eur. J. 2012, 18,
15158; c) L. Wang, X.-J. Wei, W.-L. Lei, Q. Liu, Chem. Commun. 2014,
50, 15916; d) Y.-M. Su, Y. Hou, F. Yin, X.-S. Wang, Org. Lett. 2014, 16,
2958; e) L. Wang, X.-J. Wei, Q. Liu, Org. Lett. 2014, 16, 5842; f) M.-C.
Belhomme, A. Bayle, X. Pannecoucke, Eur. J. Org. Chem. 2015, 1719.
Z. Ruan, S. Zhang, C. Zhu, P. N. Ruth, D. Stalke, L. Ackermann,
Angew. Chem. Int. Ed. 2017, 56, 2045.
[8]
[9]
a) C. Yuan, L. Zhu, Yu. Lan, Y. Zhao, Angew. Chem. Int. Ed. 2018, 57,
1277; b) C. Yuan, Y. Lan, Y. Zhao, Nat. Commun. 2018, 9, 1189.
[10] X.-G. Wang, Y. Li, Y.-M. Liang, Chem. Commun. 2018, 54, 9541.
[11] a) M. Ye, G.-L. Gao, J.-Q. Yu, J. Am. Chem. Soc. 2011, 133, 19090; b)
C.-L. Ciana, M. J. Gaunt, Angew. Chem. Int. Ed. 2011, 50, 458.
[12] a) C. F. Rosewall, F. E. Michael, J. Am. Chem. Soc. 2009, 131, 9488;
b) Y. Saito, Y. Segawa, K. Itami, J. Am. Chem. Soc. 2015, 137, 5193.
[13]
a) B. Ma, Z. Chu, L. Liu, J. Zhang, Angew. Chem. Int. Ed. 2017, 56,
2749; b) K. Raghuvanshi, D. Zell, L. Ackermann, ACS Catal. 2016, 6,
3172; c) H. Wang, J. Koeller, L. Ackermann, Chem. Eur. J. 2015, 21,
15525.
[14] a) S. Okumura, S. Tang, Y. Nakao, J. Am. Chem. Soc. 2016, 138,
14699; b) L. Yang, K. Semba, Y. Nakao, Angew. Chem. Int. Ed. 2017,
56, 4853; c) S. Okumura, Y. Nakao, Org. Lett. 2017, 19, 584.
[15] Y.-L. Xiao, X. Zhang, Angew. Chem. Int. Ed. 2014, 53, 9909.
[16] a) O. Saidi, J. Marafie, C. G. Frost, J. Am. Chem. Soc. 2011, 133,
19298; b) N. Hofmann, L. Ackermann, J. Am. Chem. Soc. 2013, 135,
5877; c) J. Li, L. Ackermann, J. Am. Chem. Soc. 2015, 137, 13894.
[17] Z. Jiao, L. H. Lim, J. S. Zhou, Angew. Chem. Int. Ed. 2018, 57, 6294.
[18] C. C. C. J. Seechurn, T. Sperger, T. J. Colacot, J. Am. Chem. Soc.
2017, 139, 5194.
Scheme 4. Plausible mechanism
In conclusion, we have developed a practical and general
strategy for the direct para-selective difluoromethylation of
aromatic ketones by employing Pd(PPh3)4 as the catalyst.
[19] H. Wang, Y. Ma, H. Tian, A. Y. Wu, Tetrahedron 2014, 70, 2669.
This article is protected by copyright. All rights reserved.

10.1002/anie.201809788
Angewandte Chemie International Edition
COMMUNICATION
[20] Y. Motoyama, Z. Zhang, S.–H. Yoon, ChemCatChem 2018, 10, 505.
[21] I. S. Kondratov, M. Y. Bugera, N. A. Tolmachova, G. G. Posternak, C.
G. Daniliuc, G. Haufe, J. Org. Chem. 2015, 80, 12258.
[22] a) D. C. Powers, T. Ritter, Nature Chemistry 2009, 1, 302; b) N. R.
Deprez, M. S. Sanford, J. Am. Chem. Soc. 2009, 131, 11234; c) Z.
Feng, Q.-Q. Min, X. zhang, Angew. Chem. Int. Ed. 2014, 53, 1669.
This article is protected by copyright. All rights reserved.

10.1002/anie.201809788
Angewandte Chemie International Edition
COMMUNICATION
Entry for the Table of Contents (Please choose one layout)
Layout 1:
ESSAY
Text for Table of Contents
Author(s), Corresponding Author(s)*
Page No. – Page No.
Title
((Insert TOC Graphic here))
Layout 2:
ESSAY
G. Tu, C. Yuan, Y. L，J. Zhang and
Prof. Dr. Y.-S. Zhao*
Page No. – Page No.
A Ligand Enabled Palladium
Catalyzed Highly para-Selective
Difluoromethylation of Aromatic
Ketones
A practical and highly para-selective C–H difluoromethylation of aromatic ketones
has been developed by employing tetrakis(triphenylphosphine)palladium(0) as the
catalyst and triphenylphosphine as the ligand. In addition to general aromatic
ketones, this transformation was compatible with bioactive compounds and well-
known drugs, such as oxybenzone, ketoprofen, zaltoprofen, and propafenone.
This article is protected by copyright. All rights reserved.